Effects of Thinning and Herbicide Application
on Vertebrate Communities in Longleaf
Pine Plantations
Kristina J. Brunjes,1 D. B. Warnell School of Forest Resources, University
of Georgia, Athens, GA 30602
Karl V. Miller, D. B. Warnell School of Forest Resources, University of
Georgia, Athens, GA 30602
W. Mark Ford, USDA Forest Service, Northeastern Research Station,
Parsons, WV 26287
Timothy B. Harrington,2 D. B. Warnell School of Forest Resources,
University of Georgia, Athens, GA 30602
M. Boyd Edwards, USDA Forest Service, Southern Research Station,
Athens, GA 30602
Abstract: Currently, nearly 98% of the land area once dominated by longleaf pine ecosys­
tems has been converted to other uses. The U.S. Forest Service is replanting logged areas
with longleaf pine at the Savannah River Site, New Ellenton, South Carolina, in an effort
to restore these ecosystems. To ascertain the effects of various silvicultural management
techniques on the vertebrate communities, we surveyed small mammal, herpetofaunal, and
avian communities in six 10- to 13-year-old longleaf pine plantations subjected to various
thinning and herbicide regimes. Areas within each plantation were randomly assigned one
of four treatments: thinning, herbicide spraying, thimling and herbicide, and an untreated
control. For all vertebrate groups, abundance and species diversity tended to be less in the
controls than treated areas. Birds and small mammals were most abundant and diverse in
thinned treatments versus spray only and control. Herpetofauna capture rates were low and,
thus, we were unable to detect treatment-related differences. Silvicultural treatments that
reduce hardwood stem density and pine basal area can enhance habitat conditions for nu­
merous vertebrate species.
Key words: avian communities, herpetofauna, longleaf pine, silvicultural treatment,
small mammals.
Proc. Aru1U. Conf. Southeast. Assoc. Fish and Wildl. Agencies 57:252-267
Nearly 98% of the land area once dominated by longleaf pine (Pinus palustris)
ecosystems has been converted to other uses such as suburban development, agri­
culture, or production of faster-growing pine species (Ware et al. 1993). This decline
1. Present address for Kristina Bnmjes: Department of Range, Wildlife, and Fisheries Manage­
ment, Texas Tech University, Lubbock, TX 79409
2 . Present address for Timothy Harrington: Pacific Northwest Research Station, USDA Forest
Service, 3625 93rd Ave. SW, Olympia, WA98512
2003 Proc. Amm. Conf. SEAFWA
Vertebrate communities in LongleafPine Plantations 253
continues, making restoration of this critically endangered ecosystem a priority for
many agencies, including the U.S. Forest Service (USFS). Historically, the area which
is now the Savannah River Site National Environmental Research Park (SRS) in the
Upper Coastal Plain of South Carolina was dominated by longleaf pine systems. By
the 1950s, when the area was acquired by the Department of Defense, most areas
which had not already been converted to agriculture or other uses were logged by
displaced residents. These areas were reforested, naturally and by planting, primarily
with loblolly pine (P. taeda; Workman and McLeod 1990). In recent decades, the
USFS at the Savannah River Forest Station has placed greater emphasis on restoration
and management of longleaf pine communities including replanting logged areas with
longleaf pine.
Many of these longleaf pine plantations on SRS have begun to reach canopy
closure. Time to canopy closure depends on site quality, pine species, planting
density, and other silvicultural treatments. Populations of many wildlife species
decline dramatically after canopy closure. Intensive silvicultural practices, such as
chemical site preparation and machine planting, may be used to regenerate longleaf
pine (Nelson et al. 1982). However, intensive management results in earlier crown
closure and subsequent exclusion of understOlY vegetation (Harrington and Edwards
1999). In the Georgia Piedmont, small mammal density declined significantly
following canopy closure in loblolly pine stands (Atkeson and Johnson 1979,
Langley and Shure 1980). Studies in other southern pine forests have documented
similar rapid declines in the abundance and diversity of songbirds (Johnson and
Landers 1982, Childers et al. 1986). These decreases in wildlife are attributed
to reductions in understory vegetation diversity and cover, and other changes in
vegetation structure as the stands mature (Johnson and Landers 1982). Thinning in
loblolly-shortleaf stands in Mississippi and Louisiana (McComb and Noble 1980)
and in loblolly stands in the Georgia Piedmont (King 1982) resulted in increased
small mammal abundance and diversity. Similarly, thinning in combination with
prescribed burning can increase wildlife forage yields in natural stands of longleaf
pine (Grelen and Enghardt 1973) and in plantations of loblolly pine (Hurst et al.
1981). Herbicides have been used to suppress competition fi'om other woody and
herbaceous plants in longleaf plantations (Nelson et al. 1982) and help restore the
wiregrass (Aristida stricta) community typically associated with longleaf pine
(Wilkins et al. 1993). Conversely, some herbicides can eliminate or reduce species
diversity of understory vegetation, which can be important food sources for wildlife
(Santillo et al. 1989).
Canopy closure results in decreased habitat suitability for herpetofauna in
longleaf pine forests (Guyer and Bailey 1993). Williams and Mullin ( 1987) found that
amphibians were uncommon in poletimber size longleaf-slash pine stands in Louisiana,
primarily due to lack of water sources and shaded cover. In addition, they found
fewer individuals and species of herpetofauna in the internlediate-aged pine stands
than in older, sawtimber-sized stands. Conversely, at the Savannah River Site (SRS),
amphibian diversity was greatest in intermediate-aged (three and eight years) loblolly
pine stands compared to mature stands (Grant et al. 1994). Intensivemanagement
associated with establishing plantations may decrease amphibian diversity by
2003 Proc. Annu. Conf. SEAFWA
254
Brunjes et at
eliminating important microhabitats. For example, plantations typically have less
coarse woody debris than unmanaged stands (deMaynadier and Hunter 1995).
In longleaf plantations, pre-commercial thinning and herbicide applications may
delay the onset of canopy closure or re-open closed stands. Several studies have
investigated the effects of thinning or herbicide treatments on particular components
of the wildlife community, such as small mammals, in southern pines. However, these
studies did not examine the combined effects of such treatments on the birds, small
mammals, and herpetofauna of young longleaf pine stands. To make management
decisions that will provide habitat conditions for a variety of wildlife species within
these plantations, it is necessary to determine if measurable changes in vertebrate
abundance and species assemblages occur within and between stands treated
mechanically and with herbicides. We compared small mammal, herpetofaunal, and
avian communities in longleaf pine plantations subjected to various thinning and
herbicide regimes over a 2-year period.
Study Area
The study was conducted within an existing research project (Harrington and
Edwards 1999) in the Sandhills region of the Upper Coastal Plain physiographic
province at the Savannah River Site near New Ellenton, South Carolina. We selected
6 longleaf plantations as study sites. All sites were within well-stocked plantations
(2: 1,500 trees/ha; 2: 60% closure estimated visually) with an intermediate to co­
'
dominant stratum of hardwoods (2: 500 trees/ha). Hardwoods consisted primarily
of sand post oak (Quercus margaretta), turkey oak (Q. laevis), water oak (Q. nigra)
and hickories (Carya spp.) saplings. At study initiation in early 1996, the plantations
ranged in age from 8 to 11 years and were relatively productive for longleaf pine (site
indexso 2: 24 m). Soils were well drained to excessively well drained and included
the Blanton, Lakeland, and Troup series.
Methods
All study sites were prescribed-burned in February-March 1994. We used a
randomized complete block design consisting of six replications and four treatments.
Treatment areas were 3-7 ha in size and included:
1) Untreated control-No other treatments were applied except the 1994 pre­
scribed burn.
2) Thin-Pines were thinned to an average stem density of 635 tree/ha. Trees
were cut and left on the ground to decay, resulting in minimal disturbance to the
litter layer and soil and increased woody debris.
3) Spray-Pines were unthinned. Hardwoods and shrubs were treated with
herbicides as described below. All dead vegetation was left standing.
4) Thin + Spray-Combination of thin and spray treatments as described
above.
Within each of the 24 treatment areas (six sites x four treatments), we marked 10
permanent sample points spaced on a 40-m by 40-m grid. Two of the six sites were
2003 Pl'Oc. Annu. Conf. SEAFWA
Vertebrate communities in LongleafPine Plantations 255
burned again in February 1997 when a prescribed fire in an adjacent mature pine
stand escaped. Consequently, data from these replicates in spring and summer of
1997 were not included in analyses.
In May-June 1994, pines in thin and thin + spray treatment areas were cut to
leave an average of 625 trees/ha (4.0 m spacing). Hardwood/shrub removal in spray
and thin + spraytreatrnent areas was initiated in March 1995 with a 1- x I-m spot­
grid application of Velpar L herbicide (hexazinone). The application rate was 1.7 kg
active ingredient (aj.)/ha, except in dense hardwood/shrub areas where it was 2.2
kg aj.lha. Surviving hardwoods and shrubs (eg., Vaccinium spp., Prunus spp., and
Carya spp.) were treated in March 1996 with a basal stem application of Garlon 4
herbicide (triclopyr, 7% in oil). During summer 1996, directed foliar applications of
Arsenal (imazapyr, 0.5%) plus Accord (glyphosate, 5%), and stem injections of the
two chemicals (5% and 50% concentrations in water, respectively) were applied to
eliminate most surviving hardwoods and shrubs.
Vegetation was sampled in August 1996 as described in Harrington and Ed­
wards ( 1999). Coverage (%) was estimated for each species using the line-intercept
method. For each sample point, crown intersections (nearest cm) per species along a
permanently-marked plot radius (3.6 m) were recorded. Percentage cover of a given
species was calculated by dividing the total length of its crown intersections by
transect length then multiplying by 100. In December 1996, diameter (cm) at breast
height ( 1.37 m) was measured on each tree >2.5 cm rooted within 6 m of a given
sample point. Height (m) was measured on a random sample of 20% of the trees.
Stand basal area (m2/ha) and stem density (N/ha) were calculated from these data.
We sampled small mammal populations by removal trapping (Jones et al. 1996)
at four of the six sites. One Victor rat-trap was placed 4 m north or south of each
sample point. One Victor mousetrap was placed opposite the rat-trap, 4 m from each
sample point. Traps were baited daily with peanut butter and oatmeal. Each trapping
period consisted of sampling all treatment plots in sites 1-4 for four consecutive
nights ( 1,280 trapnights/sampling period). Trapping was conducted in April 1996,
July-August 1996, December 1996, April 1997, and August 1997. Animals were
identified to species using morphological characteristics (Cothran et al. 1991).
Herpetofauna and soricid abundances were assessed using drift fence arrays
with pitfall traps (Kirkland and Sheppard 1994, Ford et al. 1999). We constructed
32 drift fence arrays on the same four sites where snap-trapping was conducted.
Drift fences were linear, 9m long, perpendicular to the slope, each with five 5-gallon
buckets evenly spaced along its length. A small amount of soil, litter, and water was
maintained in each bucket to provide captured animals with shelter and protection
from desiccation and fire ants. Mall1ll1als, lizards, and amphibians were toe-clipped
in a cohort-marking scheme (Dolllelly et al. 1994) and released immediately at the
point of capture. Snakes were held until the end of the trapping period and released
at the point of capture. During April 1996, July/August 1996, April 1997, and August
1997 pitfalls were opened for 10 consecutive nights.
To index abundance and diversity of breeding birds, permanent point-count
locations were established at the center of each treatment area on all six study sites.
2003 Proc. Annu. Conf. SEAFWA
256
Table 1.
Brunjes et al.
Mean (SE) size and abundance of trees (December 1996) and ground coverage
of understory vegetation and woody debris (August 1996) in longleaf pine plantations at
the Savannah River Site, South Carolina. Pines were thinned in May 1994, hardwoods
and shrubs were removed with herbicides in 1995-1996, or the combined treatments were
applied.
..
Treatment'
DBH(cm)b
Pines
Hardwoods
Height(m)
Pines
Hardwoods
Stem density (Nlha)
Pines
Hardwoods
Basal area (m2/ha)C
Pines
Hardwoods
Herb coverage (%)
Shrub coverage (%)d
Tree seedling coverage (%)d
VlIle coverage (%)d
Woody debris coverage (%)d
Spray
Control
ThIn
Thin+spray
11.8 (O.7)a
4.8 (0.3)a
10.9 (O.5)ab
10.9 (0.6)ab
8.2 (O.6)a
5.0 (O.6)a
7.7 (OA)a
8.3 (O.5)a
8.3 (O.6)a
4.8 (0.5)a
641 (57.5)b
655 (237.3)a
641 (54.1)b
1400 (74.9)a
1502 (36.6)a
803 (157.6)a
Vegetation parameter
7.3(1.0)b
1.5 (0.5)a
33.7 (0.7)a
17.9 (0.6)a
7.8 (0.3)a
8.8 (0.1)a
3.0 (O.1)a
6.2 (O.6)b
25.5 (OA)a
2.1 (O.7)c
0.1 (O.l)b
0.6 (0.2)b
1.5 (O.1)a
13.7 (1.4)a
14.4 (O.8)b
3.4 (0.2)bc
1.2 (O.1)b
1.0 (0.3)b
0.2 (O.2)b
10.5 (0.6)b
4.3 (0.3)a
14.1 (1.6)a
1.5 (O.5)a
18.3 (O.4)b
9.3 (0.7)ab
11.7 (2.0)a
5.5 (0.2)a
0.2 (O.2)b
a. Means
within a row followed by the same letter do not differ slgnlficanUy (P >0.05).
b. Stem diameter at 1.37 m above ground.
e. Total cress-sectional area of all tnle stems of bh 2:2 .5 e m.
d
d. HarrIngton and Edwards (1999)
Each point was visited twice weekly for six weeks-the last two weeks of April,
first week of May, last week of May and the first two weeks of June 1996 and 1997.
During each 5-minute count, all birds heard or seen within 50 m of the center mark­
ers were recorded (Hutto et al. 1986). Counts were conducted within four hours of
sunrise. Species were categorized as either neotropical migrants or residents, which
included year-round residents, winter residents, and short distance migrants. We
calculated Shannon diversity and richness of avian communities according to Ma­
gulTan (1988).
Vegetation coverage data were arcsine transformed prior to analysis (HalTington
and Edwards 1999). Small mammal and herpetofauna abundance data were log­
transformed to improve non-normality. We used the multivariate analysis of vari­
ance (MANOVA) procedure in SAS (SAS 1989) to test for differences in abundance
among treatments and trapping periods (season and year) for herpetofauna and small
mammals. lf the F-test from the MANOVA was significant (P < 0.05) for a species,
then we conducted means separation for treatments with Duncan's Multiple Range
Test. We used analysis of variance to test for differences and interactions in species
richness for all vertebrate groups and for avain abundance among trapping periods
and treatments and used Duncan's Multiple Range Test for means separation.
2003 Proc. Annu. Conf. SEAFWA
Vertebrate communities in LongleafPine Plantations 257
Results
Vegetation
Diameter and height of pines and hardwoods did not differ among treatments,
except that pine diameter was greater in thin only treatments compared to controls
(Table 1). Thinning increased coverage of herbs, shrubs, and woody debris, whereas
herbicide treatment decreased cover of shrubs, vines, and tree seedlings. Additional
information regarding vegetation responses can be found in Harrington and Edwards
( 1999).
Snap-trapping
We captured 2 17 mammals of nine species during 6,381 trapnights (3.4% trap
success). Oldfield mice (Peromyscus polionotus) were captured most frequently
(44% of all captures). Cotton mice (P. gossypinus) comprised 34% of all captures.
Other species captured were Eastern woodrat (Neotoma floridana, 9%), cotton rat
(Sigmodon hispidus, 5%), golden mouse (Ochrotomys nuttalli, 3%), Southern short­
tailed shrew (Blarina carolinensis, 3%), least shrew (Cryptotis parva, 1%), pine vole
(Microtus pinetorum, 1%), and Eastern harvest mouse (Reithrodontomys humulis,
0.5%).
The overall abundance of small mammals did not differ among periods (Wilks' A
= 0.01, df= 36, F=1.67, P= 0. 11) or treatments (Wilks' A= 0.02, df= 27, F= 1.21,
P = 0.37). Intraspecific differences in abundance among treatments were observed
only for oldfield mice (Table 2). In December 1996, oldfield mice were captured
more frequently (F= 4.38, df= 3, P= 0.02) in thin + spray areas than in spray only
or untreated areas. The following spring (April 1997), oldfield mice were captured
only in thin + spray areas (Table 2).
Species richness of small mammals (number of species captured) differed among
periods (F = 9.42, df= 4, P= 0.0 1) but not treatments (F = 0.44, df= 3, P= 0.73).
The mean number of species captured declined over the course of the study: spring
1996 [ = 5.25 ( 1.50)] > summer 1996 [ = 4.00 (2.00)] = winter 1996 [ = 3.25
( 1.71)] = spring 1997 [ = 3.25 (0.50)] > summer 1997 [ = 1.75 (.050)].
Pitfall Trapping
We captured 283 individuals representing 24 species during 8,960 trapnights
(Table 3). Of these, 63% were herpetofauna, 20% were shrews, and 17% were
rodents. The herpetofauna were 28% lizards (five species), 19% snakes (six species),
14% frogs (four species), and 2% salamanders (two species). Six-lined racenlllners
(Cnemidophorus sexlineatus, 14% of total captures) were the most common vertebrate
captured in pitfalls (all seasons combined), followed by southeastern crowned snakes
(Tantilla coronata, 13%) and least shrews (11%). Least shrews accounted for 57% of
shrew captures, followed by southern short-tailed shrews (30%) and south-eastern
shrews (Sorex longirostris, 13%).
Abundance of reptile and amphibians differed among periods (Wilks' A = 0.01,
df= 36, F = 1.02, P= 0.01 ) but not among treatments (Wilks' A= 0.44, df= 36, F =
0.44, P = 0.91). Soricid abundance also differed among periods (Wilks' A= 0. 18, df
2003 Proc. Annu. Conf. SEAFWA
258
Brunjes et al.
= 15, F= 2.60, P= 0.01) but not differ among treatments (Wilks' A= 0.59, df= 9, F=
0.84, P= 0.58). Four species of rodents also were captured in the pitfall traps (cotton
mouse, Eastern wood rat, oldfield mouse, and pine vole) but numbers were insufficient
for analysis. No individual species of vertebrate captured in pitfalls exhibited differ­
ences in. abundance among treatments. Species richness of herpetofauna species also
differed among periods (F= 10.61 , df= 3, P= 0.01) but not among treatments (F=
0.40, df= 3, P= 0.75). However, no trend over time was observed in the number of
herpetofauna species.
Point Counts
Thirty-five avian species were heard or observed in 1996, and 26 in 1997. In 1996,
total avian abundance was greater in thin only and thin + spray treatments than in
spray only or control areas (F= 5.00, df= 3, P:::;O.OI; Table 4). The following year,
total avian abundance and resident abundance were greater in thin + spray areas than
in any of the other 3 treatments (F = 3.76, df = 3, P= 0.0 1). In 1996, neotropical
migrants were more abundant in thin only and thin + spray areas than in untreated
areas (F = 3.93, df= 3, P= 0.03). The following year (1997), neotropical migrants
were more abundant in thin + spray areas than in any other treatment areas (F =
2.75, df= 3, P:::;O.OI). Year-round residents and winter residents were more abundant
in thin only and thin + spray areas than in controls in 1996 (F = 2.87, df= 3, P=
0.05). In 1997, residents were more abundant in thin + spray areas than any other
treatment (F= 6.50, df= 3, P :::;0 .01). Shannon diversity indices did not vary among
treatments .in 1996 (F= 0.44, df= 3, P= 0.78) or in 1997 (F= 1.21, df= 3, P=
0.33). However, species riclmess was greater in thin only and thin + spray areas
compared to untreated areas in 1996 (F= 3.70, df= 3, P= 0.03). The following year,
richness was greater in thin + spray areas than in any of the other treatments (F =
0.72, df= 3, P :::;O.OI).
Discussion
Low capture rates of herpetofauna and small mammals are typical in sandhill
habitats (Stout and Marion 1993), making treatment effects difficult to detect. W hile
our plots did sample the entire stands, the stands themselves were not large and the
surrounding forest types varied, which may have confounded treatment effects. For
all vertebrate groups, abundance and species richness tended to be less in untreated
control areas.
Though not significant across species, mammal capture rates tended to be great­
er in the thin only and thin + spray treatment areas. This trend was due in part to the
increased amount of coarse woody debris left on the ground compared to unthimled
areas (Table 1). In addition, herbaceous plants were more abundant and had greater
plant species richness in these stands (Harrington and Edwards 1999). This com­
plexity may have allowed a more diverse community of small mammals to exist
in these stands compared to more homogenous, unthimled stands (King 1982). In­
creasing the amount of woody debris by leaving cut trees and slash and minimizing
disturbance to herbaceous cover during treatments may benefit small mammals.
Oldfield mice prefer sites with sandy soils, plentiful herbaceous vegetation, and
2003 Proc. Annu. Conf. SEAFWA
N
o
o
w
'"C
O
° stS::;-:
0
0 0.. O::""'oa
Dl
>
"-<(p oq(P
-"S
o::e..0
0
"-<:
2st8..
O
.... .
§. "-<:
0 'g
n0 "
j:;
q:s,.. ">-- O (P
rl-" <
>-' "0
''''''- rl-O ..... (p
0
0" (p
o ::1.;;5
<(Jq
g;;::
st
(P +
S.
lZl ct o·
"
(/) Sf
"-<:
lZl (P :-'?d
...... (/)"-<: ct
Q
e;e;
(P !Z' -" e;o 2..
(P
'
<(t
'. (P
0"
,, ,--"'-;
(Jq lZl (P
"
@§
O
P=1
0..
(P e:...
§
o..rl::E
.
(P (P
lZl
(t
,.....
< rl- o_
\O
O'I
§l
"-<: Vl
0
(P
o t""l
P o" " ttI
(P
"
lZl
Z S
::1.
(/)
?d (P O
2..
&'5
<
(/)
(P§0..(P
Dl
ctg. rl::E
.....
g.
P"'
0"
(p
(P
0"
"
(/)
(P
rl- S
.... ·
"0 0
lZl
0S
sg(P
....
.
rl@
ct ::E ......
&
- "80..
(P
(P Vl
P"'
&'5 @S':-'
8'0..!Z' 0 >-3
lZl
lZl
@ (P
"
"-<:
rl- (P H
v
P"'
v t""l
..... lZl (p £e..
p
lZl
,..... lZl
0
&
C/:I
.... . "O
0... lZl
""'
0
a (P
(P
o "-< lZl ,..... 0
o
0
i-J:
o
&'5(DH)
1
_.
8
()
:::
n
g
t"+>
IZl
.
.
I
.
""",,"
Mean (SE) small mammal snap-trap captures per plot by treatment (80 trap nights/replicate) in longleaf pine
plantations at the Savannah River Site, South Carolina, in 1996 and 1997.
Thble 2.
'-:
Species
Season
Treatment"
Spring
1996
N=4
Thin
Thin + spray
Spray
Control
Summer Thin
Thin + spray"
1996
N=4
Spray
Control
Wmter
1996
N=4
.
Spring
1997
N=2
Oldfield
Mouse
" Cotton
mouse
3.00(1.22)
3.75(2.10)
0.75(0.48)
0.75(0.48)
2.00(0.00)
1.50(0.50)
2.00(1.08)
1.50(0.29)
1.75(0.75)
0.75(0.48)
0.50(0.29)
0.75(0.48)
.Eastern
woodrat
mouse
Otheh>
All species
combined
50(050)
0.25(0.25)
0.75(0.75)
025(025)
0.25 (0.25)
0.25(0.25) 0.25(0.25) 0.25 (0.25) 0.50(0.29) 0.50(0.50) 0.25(0.25) 0.75(0.48) 0.50(0.50)
0.75(0.48)
0.25(0.25) 0.75(0.48) 0.50(0.50)
1.00(0.58) 0.50(050) 0.25 (0.25) 4.50(1.19)
1.75(0.25)
1.75(1.11)
2.25(0.86)
2.00(0.00)
0.50(0.50)
2.50(0.50)
1.00(1.00)
1.00(1.00)
Control Golden
6.25(1.44)
5.75 (2.25)
0.50(Q.50) 3.75(1.25)
3.50(1.04)
Thin
Thin + spray
3.25(0.95)
4.50(1.55)
0.25 (0.25) 2.00(058)"
0.50(0.50)
0.75(0.48)
0.25(0.25)
0.50(0.50)
5.00(2.00)"
Control
Cotton
rat
.50(050)
0.50(0.50)
0.25(0.25)
0.25(025)
0.50(050) Summer TIrin
Thin + spray
1997
Spray
N=2
shrew
0.50(0.50)
TIrin
L 75(l.44)abc 0.25(0.25)
Thin + spray 3.25(O.63)a 1.00(0.71)
Spray
0.75(O.48)b 0.50(0.29)
Control
0.00(O.oo)b
Spray S. shott-tailed
0.50(0.50)
3.00 (l.()()
5.50(1.50)
3.50(1.50)
1.00(1.00)
0.50(0.50)
1.00(1.00)
0.50(0.50)
1.00(1.00)
1.00(1.00)
1.00(1.00)
a. Least shrew, wOodIand vole, and Eastern bam:st moose.
b. In thin and thin + spray treatments, pine basal area was reduced to approximately one-half. In thin + spray and spray treatments, hardwoods and shrubs were removed
with herbicides.
c. Within a season and species, means followed by the same letter are not different at P <
0.01.
2
::
::
;::
::::
....
.
s·
.:;
)
:::: '
§
is'
....
.
::::
t:.:l
tv
Vl
\0
tv
0
0
w
'"0
....
0
>-
'Thble3.
Mean (SE) amphibians, reptiles, and sroan mammal captures by treatment using pitfall arrays in young longleaf pine
plantations at the Savannah River Site, South Carolina, in 1996 and 1997. Pitfalls were open for 400 nights per treatment. In
1996 N = 4 sites, in 1997, N = 2 sites.
(J
Spring
0
i:l
:-+>
1996
Summer 1996
Thin +
IZl
Thin
trJ
spray
Thiu+
Spray
Untreated
All amplnDians
Southern toad
Thin
spray
J..
Spray
Untreated
1.00(0.41)
1.50(0.65)
150(0.87)
1.75(0.63)
0.75(025)
0.75(0.48)
025 <0.25)
0.75(0.48)
025(0.25)
0.75(0.48)
0.75(0.75)
050(029)
025(025)
0.25(025)
(Bufo terrestris)
E. n arrow-m outhed toad
(Gastrophryne carolinensis)
Gteen frog
(Rana clamitans)
Slimy saIrunander
0.25{025)
(Plethodon glutinosis)
Red saJamander
0.25(0.25)
(Pseudotriton TUber)
All lizards
Green anole
(Anolis carolinensis)
Fence lizard
(ScelopoTUS wululatus)
Six-lin ed racemnner
(CnemidophoTUS sexlineatus)
Ground skink
(Scincella lateralis) .
Five-lined sk:ink
(Eumeces fasciatus)
1.00(0.40)
0.75·(025)
1.50(0.65)
1.00(0.71)
0.25(025)
0.25(0.25)
0.75(0.48)
0.75(0.48)
0.75(0.25)
0.50(029)
0.75(0.75)
0.75(025)
2.25(0.48)
0.75(0.48)
3.75(1.11)
050(0.50)
0.25(025)
0.25(025)
2.00(0.41)
4.50(1.44)
tv
0\
0
025(0.25)
0.50(050)
3.50(0.96)
3.50(1.19)
0.25(0.25)
0.25(025)
0:;
=
.:::.
('t>
r;,.,
('t>
-
ThbIe 3. , Continued
Sprin l996
Thin
Thin +
spray
Spray
SUIIlJJIe[" 1996
Untreated
All snakes
Southeastem crowned snake
(Tantilla coronata)
Smooth earth snake
(Virginia valeriae)
Thin
Thin +
spray
Spray
Untreated
0.75(0.48)
1.25(0.25)
1.50(0.50)
1.00(0.00)
0.25(025)
1.25(025)
1.50( :50)
0.75{025)
0.25(025)
::t
.
S·
t""I
0.25(025)
Scarlet kingsnake
;:
:: ;::
;:::
(Lampropeltis triangulum)
Mud snake
025(025)
(Faranda abacura)
-:;
Allsruews
0.50(0.50)
0.50 (0.50)
1.00 (0.41)
2.00(1.08)
1.50(0.65)
1.50(029)
Least shrew
(Cryptotis parva)
025(025)
0.50(0.50)
025(025)
125(0.95)
0.50(029)
1.00(0.00)
025(025)
025(0.25)
1.00(0.41)
0.50(029)
0.50(0.29)
0.50(0.29)
S. short-tailed shrew
(Blarina carolinensis)
Southeastem shrew
N
0
0
w
"1:1
(Sora longirostris)
S'
Oldfield mouse
(Peromyscus poliolWtus)
Cotton mouse
(P. gossypinuS)
Woodland vole
(Microtus pinetorum)
'"'
0
;::
F
n
0
i:l
t"'>
en
trI
All rodents
E. woodrat
(Neotomafloridana)
0.25(025)
;::
(I:>
'
S"
;::
:;::-.
;::
'"
0.25(0.25)
0.25(0.25)
0.25(0.25)
025(0.25)
2.00(1.41)
2.50(0.87)
0.50 029)
125(0.48)
025(025)
225(0.63)
050(029)
050(029)
0.50(029)
125 (0.95)
0.25(025)·.
0.50(050)
N
0\
.....
N
0
0
<.>.l
'"d
....
0
(>
Table 3.
Continued
Spring 1997
>-
Thin
()
0
All amphibians
E. narrow-mouthed toad
C/J
Bullfrog
;:!
t+>
Thin +
spray
Spray
LOO(0.00)
Summer 1997
Untreated
0.50(0.50)
Thin
0.50(0.50)
0.50(0.50)
Thin +
spray
1.50(1.50)
1.50(1.50)
S pray
U
..
(Gastrophryne carolinensis)
0.50(0.50)
(Rana catesbiana)
Green frog
(Rana clamitans)
Red salamander
050(0.50)
(Pseudotriton TUber)
All lizards
Greenanole
0.50(0.50)
2.00(2.00)
1.50(1.50)
(Arwlis carolinensis)
Six-lined racenmner
(Cnemidophorus sexlineatus)
All snakes
Southeastern crowned snake
0.50(0.50) (Tantilla coronata)
Common garter snake
2.00(1.00)
1.00(0.00)
050(0.50)
0.50 (0.50)
1.00(0.00)
050(050)
0.50(050)
2.00(1.00)
0.50(0.50)
(Thamnophis sirtalis)
Smooth earth snake
1.00(0.00)
2.00(0.50)
050 0.50)
050(050)
0.50(0.50)
(Virginia vaIeriae)
BIackracer
0.50(0.50)
0.50(050)
1.00(1.00)
050(0.50)
1.50(050)
0.50(050)
0.50(0.50)
1.00(1.00)
0.50(050)
1.00(1.00)
0.50(0.50)
(Coluber constrictor)
All shrews
Leastsbrew
CO
(Cryptotis parva)
S. short-tailed shrew
(Blarina caroliriensis)
tv
0'\
tv
0.50(050)
0.50(050)
'"I
=
==.
....
(!)
IZl
(!)
-
=
:-
Table 3.
Continued
Spring 1997
Thin
All rodents
Oldfield mouse
(Peromyscus polionotus)
Cotton mouse
(P. gossypinus).
Woodland vole
(Microtus pinetorum)
Thin +
spray
Spray'
050(050)
050(0.50)
Summer 1997
Untreated
Thin +
Thin
spray
1.00 (0.00)
0.50 (0.50)
050(050)
Spray
::::
Untreated
050(050)
0.50(0.50)
050(0.50)
050(050)
(":,
::
:::!
::
.....
:::t-.
S·
.
§
.....
tv
o
o
w
>-0
.....
o
>­
::l
(J
o
i:l
(Il
.
tv
0\
W
264
Table 4.
Brunjes et al.
Mean (SE) breeding season avian abundance, diversity, and richness by treatment
in longleaf pine plantations at the Savannah River Site, South Carolina, in spring 1996 and
1997. Abundance values represent mean number of birds per survey. '!\vo sites were elimi-
nated from analysis in 1997 due to an accidental burn .
.,
Treatments
...
Migration Strategy
1996 (N = 6 sites )
Avian abundance
Neotropica1 migrant
Year-round or winter
resident
Total abundance
Shannon diversity (H')
Species richnessb
1997 (N =4 sites )
Avian abundance
Neotropica1 migrant
Year-round or winter
resident
. Total abundance
Shannon diversity (H')
Species richness
Thin
Thin + spray
Spray
Control
0.40 (O.09)8a
1.38 (O.17)a
0.42 (O.ll)a
1.60 (O.21)a
0.19 (0.05)00
1.08 (0.17)ab
0.15 (0.07)b
0.94 (0.16)b
1.77 (0.20)a
1.99 (0.09)
1.65 (O.1S)a
2m (O.25)a
2.13 (0.11)
1.S3 (0.22)a
1.23 (0.17) b
1.94 (0.12)
1.09 (0.13)ab
1.13 (0.18)b
1.91 (O.OS)
1.08 (0.16)b
0.19 (0.07)b
1.02 (0.16)b
0.44 (0.12)a
1.88 (0.2S)a
0.15 (O.OS)b
1.06 (0.16)b
0.04 (0.03)b
0.79 (0.12)b
1.21 (0.18)b
1.46 (0.16)
0.98 (0.13)b
2.31 (0.30)a
1.66 (0.19)
1.79 (0.25)a
1.21 (0.17)b
1.33 (0.22)
1.13 (O.1S)b
0.83 (0.12)b
1.52 (0.12)
0.81 (O.l1)b
a.Within a row,means followed by the same letter do not differ at P :so 0.05. b.Mean number of species per survey. curs (Golley et al. 1965), we captured only 10 individuals. Prior to canopy closure
in loblolly pine plantations cotton rats account for up to 90% of total small mammal
captures (Atkeson and Johnson 1979, Mengak et al. 1989). The low capture rates in
our study may have been a result of the lack of heavy herbaceous cover prefelTed
by cotton rats, the relatively small size of the treatment areas (Yates et al. 1997), or
reduction in habitat suitability of pine stands over time.
All of our sites were characterized by well-drained soils and were not close to
permanent water sources, making them generally unsuitable habitat for amphibians
regardless of treatment (Williams and Mullins 1987). However, southem toads and
eastem narrow-mouthed toads, common, ubiquitous species on the SRS (Gibbons
and Semlitsch 199 1) were captured on our sites during both summers. Species di­
versity (eight) was less than that in 8-year old loblolly plantations ( 15 species) on
SRS (Grant et al. 1994).
Reptiles were more numerous on the study area than amphibians, particularly
six-lined racerunners and southeastem crowned snakes. Abundance and diversity
did not vary by treatment, however, whether this was due to insufficient captures
or lack of treatment effects is unknown. Maintenance of early successional stages,
rather than the specific method of maintenance used may be more important for
reptiles in pine habitats (Greenberg et al. 1994).
2003 Proc. Annu. Conf. SEAFWA
Vertebrate communities in LongleafPine Plantations 265
In both years, avian abundance and species richness were greatest in thin +
spray areas, which had less vegetation coverage at the canopy and midstory levels.
Abundance and richness were least in untreated stands for neotropical migrants and
residents, possibly because these stands have the greatest canopy cover. Further­
more, none of our stands had more than three large snags (>20 cm dbh) within any
treatment area and mast had no snags. Presence of snags positively influences bird
abundance in early-successional habitats (Johnson and Landers 1982, Childers et al.
1986) and managers should leave standing dead trees or create snags in these planta­
tions during future thinning operations.
Precommercial thinning benefits small mammals and avifauna by increasing
herbaceous ground cover and the amount of coarse woody debris. However, pine
thinning was a less selective method than herbicide application to modify abun­
dance of specific plant species with greater value as wildlife forage. Herbicides can
be used to improve forage availability for favored species of wildlife (McComb and
Hurst 1987) and may increase stmctural diversity by leaving standing dead shmbs
and saplings within the understory. Individual herbicides have specific selectivity,
there-fore plant species composition and abundance vary with the chemical used. In
this study, three herbicides were used to virtually eliminate all non-pine woody veg­
etation. Thinning opened the canopy and increased coarse woody debris, offsetting
negative effects of herbicides without thinning seen in spray-only areas, thus result­
ing in greater species diversity and abundance of vertebrates in thin + spray treated
stands. In our study, using multiple herbicides was necessary because of the need to
control all deciduous and evergreen species that comprised the understory for other
research being conducted on our sites (Harrington and Edwards 1999). However,
other researchers have successfully established longleaf pines using only one or two
herbicide treatments (Nelson et al. 1982). Use of fewer herbicides may be desir­
able on SRS to improve habitat for wildlife during reestablishment and throughout
the rotation. Our results suggest that periodic thinning, herbicide application, and
prescribed burning can increase diversity and abundance of wildlife by delaying
canopy-closure in young plantations. Additional research is needed to determine
what management is needed to maintain a diverse wildlife community within long­
leaf plantations as they age.
Acknowledgments
Funding was provided by the U.S. Department of Energy-Savannah River Oper­
ations office through the U.S. Forest Service Savannah River and the Forest Service
Southern Research Station under Interagency Agreement DE-IA09-76SR00056.
Funding and logistical support were also provided by the University of Georgia
Daniel B. Warrell School of Forest Resources and McIntire-Stennis Proj. GEO­
0093-MS. We greatly appreciate the support provided by Dr. John Blake. We also
extend our thanks to J. H. Brunjes IV, L. A. Lewis, S. B. Castleberry, J. C. Kilgo, A.
E. Carraway, A. L. Wood, P. B. Spivey and V. A. Sparling for assistance with field
work and J. W. Gassett for assistance with statistical analyses.
2003 Proc. AlillU. Conf. SEAFWA
Vertebrate communities in LongleafPine Plantations 267
Jones, C., W.J. McShea, M.J. Conroy, and T.H. Kunz. 1996. Capturing mammals. Pages
1 15- 156 in D. E. Wilson, F. R. Cole, J. D. Nichols, R. Rudran, and M. S.Foster, edi­
tors. Measuring and monitoring biological diversity: standard methods for mammals.
Smithsonian Institution Press, Washington, DC.
King, KJ. 1982. Effects of prescribed burning and thinning in natural Piedmont 10b101­
lyshortleaf pine stands on small mammal populations. T hesis. University of Georgia,
Athens.
Kirkland, G.L. and P.K Sheppard. 1994. Proposed standard protocol for sampling small
mammals communities. Pages 277-284 in Advances in the biology of shrews, J. F.
Merritt, editors. Carnegie Museum of Natural History Special Publication 18.
Langley, AK, Jr. and D.J. Shure. 1980. The effects of loblolly pine plantations on small
mammal populations. American Midland Naturalist 103:59-65.
Magurran, A. E. 1988. Ecological diversity and its measurement. Princeton University
Press, Princeton, New Jersey.
McComb, W.C. and G.A Hurst. 1987. Herbicides and wildlife in southern forests. Pages
28-29 in J. G. Dickson, and O. E. Maughn, editors. Managing southern forests for
wildlife and fish. USDAForest Service General Technical Report SO-65.
___
and R.E. Noble. 1980. Small mammals and bird use of some unmanaged and man­
aged forest stands in the mid-south. Proceedings of the Annual Conference of the South­
eastern Association ofFish and Wildlife Agencies 34:482-49 1 .
Mengak, M.T., D.C. GUYilll, Jr., and D.H. Van Lear. 1989. Ecological implications of lob101­
1ypine regeneration for small mammal communities.Forest Science 35:503-514. Nelson, L.R., S.A Knowe, and D.H. Gjerstad. 1982. Use of effective herbicide treatments for artificial regeneration of long1eaf pine. GeorgiaForest Research Paper 40.
Santillo, D.J., D.M. Leslie, Jr., and P.W. Brown. 1989. Responses of small mammals and
habitat to glyposate application on c1em'cuts. Journal of Wildlife Management 53: 164­
172.
SAS 1989. SAS/Stat user's guide, version 6. Volume 2. 4th edition SAS Institute Inc., Cary,
North Carolina.
Stout, 1.J. and W.R. Marion. 1993. Pine flatwoods and xeric pine forests of the southern
(lower) coastal plain. Pages 373-446 in W.H. Martin, S. G. Boyce, and A C. Echter­
nacht, editors. Biodiversity of the southeastern United States: Lowland terrestrial com­
munities. John Wiley and Sons, New York, New York.
Ware, S., C.Frost, and P. Doerr. 1993. Southern mixed hardwood forest: The former 10ng­
leaf pine forest. Pages 447-493 in W. H. Martin, S. G. Boyce, and A C. Echternacht,
editors. Biodiversity of the southeastern United States: Lowland terrestrial communi­
ties. John Wiley and Sons, New York, New York.
Wilkins, R.N., G.T. Tanner, R. Mulholland, and D.G. Neary. 1993. Use of hexazinone for
understory restoration of a successionally-advanced xeric sandhill inFlorida. Ecologi­
cal Engineering 2:3 1-48.
Williams, KL. and K Mullin. 1987. Amphibians and reptiles of longleaf-slash pine stands
in central Louisiana. Pages 1 16- 1 20 in Proceedings of the southern evaluation project
workshop. USDAForest Service General Technical Report SO-68.
Workman, S.W. and KW. McLeod. 1990. Vegetation of the Savannah River Site: major
community types. National Environmental Research Park Program SRO-NERP-2 1 .
Yates, M.D., S.C. Loeb, and D.C. Guynn, Jr. 1997. T he effect of habitat patch size on small
mammal populations. Proceedings of the Annual Conference of the Southeastern As­
sociation ofFish and W ildlife Agencies 5 1 :50 1-5 10.
2003 Proc. Annu. Conf. SEAFWA
About This
File:
pnnted pUblication.
identified by the soft'Nare have been corr9'cted;
however, sonre nils takes o1ay relnaln.
ThiS file was created by scanning lhe
Misscans
266
Brunjes et al. Literature Cited
Atkeson, T.D. and AS. Johnson. 1979. Succession of small mammals on pine plantations in
the Georgia Piedmont. American Midland Naturalist 101:385-392.
Briese, L. A. and M. H. Smith. 1974. Seasonal abundance and movement of nine species of
small manunals. Journal of Mammalogy 55:615-629.
Childers, E.L., T.L. Sharik, and C.S. Adkisson. 1986. Effects of loblolly pine plantations a
songbird dynamics in the Virginia Piedmont. Journal of Wildlife Managemen 50:406­
413.
Cothran, E.G., M.H. Smith, 1.0. Wolff, and 1.B. Gentry. 1991. Mammals of the Savannah
River Site. National Environmental Research Park Program SRO-NERP-21.
deMaynadier, P.G. and M.L. Hunter, Jr. 1995. The relationship between forest managemer,
and amphibian ecology: a review of the North American literature. Environmental Re­
view 3:230-261.
Donnelly, M. A, C. Guyer, J. E. Juterbock, and R A Alford. 1994. Techniques for marking
amphibians. Pages 277-284 in W. R Heyer, M. A. Donnelly, R W. McDiannid, L. C
Hayek, and M. S.Foster, editors. Measuring and monitoring biological diversity: stan­
dards and methods for amphibians. Smithsonian Institution Press, Washington, DC.
Ford, W. M., M. A Menzel, D. W. McGill, 1. Laenl1, and T. S. McCay. 1999. Effects of a
conununity restoration fire on small mammals and herpetofauna in the southern Appa­
lachians.Forest Ecology and Management 114:233-243.
Gibbons, 1.w. and RS. Semlitsch. 1991. Guide to the reptiles and amphibians of the Sa­
vanna River Site. University of Georgia Press, Athens.
Golley, F.B., 1.B. GentlY, L.D. Caldwell, and L.B. Davenport, Jr. 1965. Number and va­
riety of small malmnals on the AEC Savannah River Plant. Journal of Mammalogy
46:1-18.
Grant, B.W. K.L. Brown, G.W.Ferguson, and J.W. Gibbons. 1994. Changes in amphibian
biodiversity associated with 25 years of pine forest regeneration: implications for bio­
diversity management. Pages 354-367 in S.K. Majumdar, F. 1. Brenner, 1. E. Lovich,
1. Shalles, and E. W. Miller, editors. Biological diversity: problems and challenges.
Pennsylvania Academy of Science, Philadelphia.
Greenberg, C.H., D.G. Nealy, and L.D. Harris. 1994. Effect of high-intensity wildfire and
silvicultural treatments on reptile conununities in sand-pine scrub. Conservation BioI
8:1047-1057.
Gre1en, H.E. and H.G. Enghardt. 1973. Burning and thinning maintain forage in a longleaf
pine plantation. Journal ofForestry 71: 419-425.
Guyer, C. and M.A. Bailey. 1993. Amphibians and reptiles of longleaf pine cOlnnmnities.
Proceedings of the Annual Tall Timbers Fire Ecology Conference 18 :139-148.
Harrington, T.B. and M.B. Edwards. 1999. UnderstOlY vegetation, resource availability, lit­
terfall responses to pine thinning and woody vegetation control in longleaf pine planta­
tions. Canadian Journal ofForestly Research 29:1055-1064.
Hurst, G.A, J.1. Campo, and M.B. Brooks. 1981. Deer forage in a burned and burned­
thinned pine plantation. Proceedings of the Annual Conference of the Southeast. As­
sociation ofFish and Wildlife Agencies 34:476-481.
Hutto, RL., S.M. Pletschet, and P. Hendricks. 1986. A fixed-radius point count meth non­
breeding and breeding season use. Auk 103:593-602.
Johnson, A.S. and 1.L. Landers. 1982. Habitat relationships of surnnler resident birds in pine
flatwoods. Journal of Wildlife Management 46:416-428.
2003 Proc. Annu. Conf. SEAFWA
About this file: This file was created by scanning the printed publication. Some mistakes introduced by scanning may remain.